IAES International Journal of Artificial Intelligence (IJ-AI)
Not a member yet
1769 research outputs found
Sort by
Leveraging machine learning techniques for student’s attention detection: a review
Get PDFWith the advances of the internet and today's innovation, it has become conceivable to conduct teaching and learning activities remotely through the online platform. Existing research says that student’s attention state and learning result are strongly correlated. However, despite its importance, this can be a challenging task, as students in general taking an online class may be in a variety of different environments and may be multitasking or distracted by other factors. This review paper aims to address these challenges by exploring the opportunities offered by machine learning techniques in attention detection for effective online teaching and learning. By leveraging machine learning algorithms, which can analyze large volumes of data, including eye-tracking, facial expressions, and body movements, we can develop robust models for attention detection in online learning environments. This paper reviews the challenges specific to online learning, such as students' attention deficits and learning styles, and highlights the limitations of current attention detection methods. Furthermore, it provides recommendations to advance attention detection technology, emphasizing the potential of machine learning to enhance attention detection technology for effective online teaching and learning
An enhanced domain ontology model of database course in computing curricula
Get PDFThe ACM/IEEE Computing Curricula 2020 includes the study of relational databases in four of its six disciplines. However, a domain ontology model of multidisciplinary database course does not exist. Therefore, the current study aims to build a domain ontology model for the multidisciplinary database course. The research process comprises three phases: a review of database course contents based on the ACM/IEEE Computing Curricula 2020, a literature review of relevant domain ontology models, and a design research phase using the NeOn methodology framework. The ontology building involves the ontology reuse and reengineering of existing models, along with the construction of some classes from a non-ontological resource. The approach to ontology reuse and reengineering demonstrates ontology reusability. The final domain ontology model is then evaluated using two ontology syntactic metrics: Relationship Richness and Information Richness. These metrics reflect the diversity of relationships and the breadth of knowledge in the model, respectively. In conclusion, the current research contributes to the Computing Curricula by providing an ontology model for a multidisciplinary database course. The model, developed through ontology reuse and reengineering and the integration of non-ontological resources, exhibits more diverse relationships and represents a broader range of knowledge
Photo-realistic photo synthesis using improved conditional generative adversarial networks
Get PDFThere are a wide range of potential uses for both the forward (generating face drawings from actual images) and backward (generating photos from synthetic face sketches). However, photo/sketch synthesis is still a difficult problem to solve because of the distinct differences between photos and sketches. Existing frameworks often struggle to acquire a strong mapping among the geometry of drawing and its corresponding photo-realistic pictures because of the little amount of paired sketch-photo training data available. In this study, we adopt the perspective that this is an image-to-image translation issue and investigate the usage of the well-known enhanced pix2pix generative adversarial networks (GANs) to generate high-quality photo-realistic pictures from drawings; we make use of three distinct datasets. While recent GAN-based approaches have shown promise in image translation, they still struggle to produce high-resolution, photorealistic pictures. This technique uses supervised learning to train the generator's hidden layers to produce low-resolution pictures initially, then uses the network's implicit refinement to produce high-resolution images. Extensive tests on three sketch-photo datasets (two publicly accessible and one we produced) are used to evaluate. Our solution outperforms existing image translation techniques by producing more photorealistic visuals with a peak signal-to-noise ratio of 59.85% and pixel accuracy of 82.7%.
Framework for content server placement using integrated learning in content delivery network
Get PDFContent placement is a significant concern in content delivery networks (CDN), irrespective of various evolving studies. Existing methodologies showcase various significant unaddressed issues concerning content placement approaches' complexities. Therefore, the proposed study presents a novel computational framework towards dynamic content placement strategy using a novel integrated machine learning approach. Simplified mathematical modelling is used to formulate and solve the content placement problem. At the same time, reinforcement learning and the sequential attentional neural network have been utilized to optimize the decision-making towards placement of content servers. Designed and assessed over a Python environment, the proposed scheme is witnessed to exhibit 35% reduced bandwidth utilization, 20% reduced delay, 23% reduced computational resource utilization, and 28% reduced algorithm processing time in contrast to existing predictive content placement schemes
Classification of Tri Pramana learning activities in virtual reality environment using convolutional neural network
Get PDFTri Pramana as the local genius of Balinese society, is now adopted in the education system. This adaptation results in a Learning Cycle Model which essentially consists of three classes namely Sabda Pramana (theoretical study), Pratyaksa Pramana (direct observation), and Anumana Pramana (practicum). In learning activities, it is difficult for educators to fully observe individuals to find out the most suitable learning model. Through Virtual Environment Technology, educators can observe students more freely through the recording of students' activities. However, in its implementation, manual analysis requires large resources. Deep Learning approach based on Convolutional Neural Network (CNN) is able to automate this analysis process through the classification ability of the image of the recorded learner activity. To produce a robust CNN model, this research compares four of the most commonly used architectures, namely ResNet-50, MobileNetV2, InceptionV3, and Xception. Each architecture is tuned using a combination of learning rate and batch size. Through a 512 x 512 resolution dataset with 70% training subset (4,541 images), 20% validation (1,296 images), and 10% test (652 images), the best ResNet model is obtained with a learning rate configuration of 1e-3 and batch size 64 with an accuracy of 99.39%, precision of 99.37%, and recall of 99.42%
Insights of learning approach towards determination of potentially objectional communication in social networking
Get PDFOver the last decade, sentiment analysis has evolved significantly towards extracting the contextual knowledge associated with the communication exchanged in social networks. Irrespective of various approaches to natural language processing and constantly evolving machine learning, sentiment analysis has inherent shortcomings, which further act as an obstacle to determining hateful and offensive speech exchanged in social networks. Therefore, this paper offers a compact yet granular insight into the effectiveness of existing sentiment analysis approaches used distinctly for determining hateful and offensive speech with particular emphasis on machine learning-based methodologies. The paper further contributes towards research trend analysis followed by distinct highlights of the research gap. The paper offers a learning outcome that significantly benefits future researchers investigating the same field
EASESUM: an online abstractive and extractive text summarizer using deep learning technique
Get PDFLarge volumes of information are generated daily, making it challenging to manage such information. This is due to redundancy and the type of data available, most of which needs to be more structured and increases the amount of search time. Text summarization systems are considered a real solution to this vast amount of data because they are used for document compression and reduction. Text summarization keeps the relevant information and eliminates the text's non-relevant parts. This study uses two types of summarizers: Extractive Text summarizers and Abstractive text summarizers. The Text Rank Algorithm was used to implement the Extractive summarizer, while Bi-directional Recurrent Neural Network (RNN) was used to implement the Abstractive text summarizer. To improve the quality of summaries produced, word embedding was also used. For the evaluation of the summarizers, the ROUGE evaluation system was used. ROUGE contrasts summaries created by hand versus those created automatically. ROUGE examination of the produced summary revealed the superiority of human-produced summaries over those generated automatically. For this paper, a summarizer was implemented as a Web Application. The average ROUGE recall score ranging from 30.00 to 60.00 for abstractive summarizer and 0.75 to 0.82 for extractive text showed an encouraging result
Personalized E-commerce based recommendation systems using deep-learning techniques
Get PDFAs technology is surpassing each day, with the variation of personalized drifts relevant to the explicit behavior of users using the internet. Recommendation systems use predictive mechanisms like predicting a rating that a customer could give on a specific item. This establishes a ranked list of items according to the preferences each user makes concerning exhibiting personalized recommendations. The existing recommendation techniques are efficient in systematically creating recommendation techniques. This approach encounters many challenges such as determining the accuracy, scalability, and data sparsity. Recently deep learning attains significant research to enhance the performance to improvise feature specification in learning the efficiency of retrieving the necessary information as well as a recommendation system approach. Here, we provide a thorough review of the deep-learning mechanism focused on the learning-rates-based prediction approach modeled to articulate the widespread summary for the state-of-art techniques. The novel techniques ensure the incorporation of innovative perspectives to pertain to the unique and exciting growth in this field
HybridTransferNet: soil image classification through comprehensive evaluation for crop suggestion
Get PDFSoil image classification is a critical task within the realms of agriculture and environmental applications. In recent years, the integration of deep learning has sparked significant interest in image-based soil classification. Transfer learning, a well-established technique in image classification, involves finetuning a pre-trained model on a specific dataset. However, conventional transfer learning methods typically focus solely on fine-tuning the final layer of the pre-trained model, which may not suffice to attain high performance on a new task. HybridTransferNet, a unique hybrid transfer learning approach designed for soil classification based on images is proposed in this paper. HybridTransferNet goes beyond the conventional approach by finetuning not only the final layer but also a select number of earlier layers in a pre-trained ResNet50 model. This extension results in substantially enhanced ability to classify when compared to standard transfer learning methods. Our evaluation of HybridTransferNet, conducted on a soil classification dataset, encompasses the reporting of various performance indicators, such as the F1 score, recall, accuracy, and precision. Our findings from experiments highlight HybridTransferNet's advantages over conventional transfer learning strategies, establishing it as a state-of-the-art solution in the domain of soil classification
Deep learning-based prediction of float model performance in floatplanes: A case study on lift-to-drag coefficient ratio
Get PDFDeveloping an engineering design is resource-intensive and time-consuming, particularly for the floats of a floatplane design, due to its complexity and limited testing facilities. Intelligent-based computational design (IBCD) techniques, which integrate computational design techniques and machine learning (ML) algorithms, offer a solution to reduce required testing by providing predictions. This paper proposes a deep learning (DL)-based IBCD method for modeling floats' lift-to-drag coefficient ratio (CL/CD), where DL is one of the most powerful ML. The proposed method consists of two phases: hyper-parameter optimization and DL model training and evaluation. A genetic algorithm (GA) is employed in the first phase to explore complex hyper-parameter combinations efficiently. Evaluation of the predicted CL/CD of the floats using the DL model resulted in a satisfactory R-squared of 0.9329 and the lowest mean squared error (MSE) of 0,001536. These results demonstrate the ability of DL model to predict the float's performance accurately and can facilitate further design optimization. Thus, the proposed method can offer a time-efficient and cost-effective solution for predicting float performance, aiding in optimizing floatplane designs and enhancing their functionalities